Measurements of cervical immunity are important for evaluating immune responses to infections of the cervix and to vaccines for preventing those infections. Three ophthalmic sponges, Weck-Cel, Ultracell, and Merocel, were loaded in vitro with interleukin-1 (IL-1), IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12, IL-15, IL-18, gamma interferon (IFN-␥), granulocyte-macrophage colony-stimulating factor (GM-CSF), immunoglobulin A (IgA), or IgG, and sponges were extracted and evaluated for total recovery by enzyme-linked immunosorbent assay (ELISA). There was excellent (>75%) recovery for all immune markers from all three devices except for IL-6, which was poorly recovered (<60%) for all sponge types, IFN-␥, which was poorly recovered from both Weck-Cel and Ultracell sponges but was completely recovered from Merocel sponges, and IL-4, which was poorly recovered from Weck-Cel sponges but was completely recovered from Ultracell or Merocel sponges. We then compared the absolute recovery of selected markers (IL-10, IL-12, IgG, and IgA) from cervical secretion specimens collected from women using each type of sponge. There were no significant differences in the recoveries of IL-10, IL-12, and IgG from cervical specimens collected by any type of ophthalmic sponge, but there was reduced IgA recovery from Merocel sponges. However, the variability in these measurements attributable to sponge types (1 to 3%) was much less than was attributable to individuals (45 to 72%), suggesting that differences in sponge type contribute only in a minor way to these measurements. We infer from our data that the three collection devices are adequate for the measurements of IL-1, IL-2, IL-5, IL-12, IL-15, IL-18, and IgG. Merocel may be a better ophthalmic sponge for the collection of cervical secretions and measurements of IL-4, IL-8, IL-10, GM-CSF, and IFN-␥, but our data from clinical specimens, not in vitro-loaded sponges, suggested the possibility of reduced recovery of IgA. These findings require confirmation.We have been interested in optimizing the collection of cervical secretions by using ophthalmic sponges to evaluate cervical immunity in the context of host response to and vaccination against human papillomavirus (HPV) infection (3-5, 7, 8, 11-13, 15). The measurement of immunity at the cervix may be necessary to understand immune responses to HPV, given the localized nature of infection and the poor correlation between levels of immunity in blood and cervical secretions (3, 4). There are numerous methods for sampling genital tract secretions, including such methods as cervicovaginal lavage and passive absorption. However, unlike cervicovaginal lavages that represent an admixture of secretions of the cervix and vagina, possibly two immunologically distinct anatomical sites (6,14), this method, like Sno-Strips wicks (Akron, Abita Springs, La.) (19), collects undiluted cervical secretions. Previously, we have shown that the collection of cervical secretions by using a Weck-Cel (XOMED Surgical Products, Inc., Jacksonville, Fla.) ophtha...